Current smokers have been the focus of recently completed, large-scale lung cancer chemoprevention trials, and the outcomes of these trials have been uniformly disappointing. There is an urgent need to understand the biologic effects of cigarette smoke on the bronchial epithelium in order to develop more effective lung cancer chemopreventive agents. Our findings indicate that bronchial metaplasia and dysplasia, which are forerunners of NSCLC, persist after smoking cessation, and we have detected clonal genetic abnormalities in histologically abnormal and normal areas of mucosa in former smokers. We have shown that the biology of the bronchial epithelium differs greatly in current and former smokers, and we hypothesize that these biologic differences are important determinants of response and resistance to lung cancer chemopreventive agents. To test this hypothesis we will perform a randomized, placebo-controlled biomarker-based clinical trial in current and former smokers, examining the effect of cyclooxygenase-2 (COX-2) inhibition on biomarkers of lung cancer risk and COX-2-dependent signaling pathways. Specific Aim 1 Hypothesis: Current and former smokers respond to the chemopreventive effects of COX-2 inhibitors, namely celecoxib. Aim: To examine the effect of celecoxib treatment on Ki-67 expression, a marker of cell proliferation, in the bronchial epithelium of current and former smokers. Specific Aim 2 Hypothesis: Current and former smokers tolerate treatment with COX-2 inhibitors. Aim: To examine the toxicity associated with celecoxib administration. Specific Aim 3 Hypothesis: Treatment with COX-2 inhibitors modulates COX-2- and lipoxygenase (LOX)-dependent signaling in the bronchial epithelium. Aim: To measure the effect of celecoxib treatment on arachidonic acid metabolites in the bronchial epithelium of current and former smokers.